WO2002001489A1

WO2002001489A1 - Segmentation of a postal object digital image by hough transform

Info

Publication number: WO2002001489A1
Application number: PCT/FR2001/001902
Authority: WO
Inventors: Gilles Arcas-Luque; Olivier Pietri
Original assignee: Solystic
Priority date: 2000-06-27
Filing date: 2001-06-19
Publication date: 2002-01-03
Also published as: PT1301896E; IL151682A; NO20026228L; ES2252251T3; EP1301896A1; CN1423794A; EP1301896B1; BR0111955A; JP2004502244A; CA2405082A1; US7110568B2; DE60116876T2; AU2001269186B2; DE60116876D1; US20040005081A1; NO20026228D0; DK1301896T3; FR2810765A1; ATE316669T1; FR2810765B1

Abstract

The invention concerns a method for segmenting a postal package digital image for automatically reading a postal address (AP) on said package, which consists in dividing (20) the image into image blocks of identical size, applying (30) on each image block a processing based on a Hough transform to identify in said image block a preferred direction of the distribution of image points in said block concerned, assembling together (40) image blocks adjacent in the image and having a common preferred direction to define (50) a rectangular zone of interest wherein is inscribed the assembly of said adjacent blocks so that said rectangular zone of interest has a longitudinal edge parallel to said common preferred direction, and applying (60) an algorithm for automatic recognition of addresses on said zone of interest of the image so as to automatically read a postal address on the postal package.

Description

The invention relates to a method of processing a digital image of a postal object in order to automatically read a postal address on said object by applying a algorithm for automatic recognition of addresses on said image.

Such a method is more particularly intended for the automatic reading of a postal address on postal packets in an automatic sorting process of postal packets. The automatic reading of addresses on postal objects of the letter type is conventionally carried out by conveying on edge of each letter to pass it in front of a camera so as to capture a digital image of the letter which can then be binarized and filtered. Generally, the lines of characters constituting a postal address to be read automatically on a letter are arranged parallel to the foot of the letter or perpendicular to it, that is to say parallel to an edge of the image, so that a Automatic address recognition algorithm can be applied directly to a rectangular part of the digital image by decoding the characters in the image in the direction given by an edge of the image. Postal items of the packet type are generally bulky items which are conveyed flat on a carpet. With this type of conveyance and this type of postal object, it is no longer possible to automatically read a postal address in a digital image of a package based on a preferred direction of the address lines provided by a predetermined reference. in the picture.

The object of the invention is to propose a solution to the problem of processing digital images of postal packets with a view to automatically reading a postal address on each packet.

To this end, the subject of the invention is a method for processing a digital image of a postal object with a view to automatically reading a postal address on said object by application of an automatic address recognition algorithm on said image. , characterized in that it consists in dividing the image into image blocks of identical size, in applying to each image block a treatment based on a Hough transformation to identify in the image block considered a direction privileged distribution of the image points in said block considered, to group together image blocks contiguous in the image and having a common privileged direction to define a rectangular area of interest in which the grouping of said contiguous blocks having a common privileged direction is inscribed so that this rectangular area of interest has an edge longitudinal parallel to said common preferred direction, and to apply the automatic address recognition algorithm to said area of interest of the image with a view to automatically reading a postal address.

With this method, a rectangular area of interest is obtained which extends longitudinally parallel to the character lines of a postal address so that it is possible by following the longitudinal edge of this area of interest to apply an algorithm automatic address recognition on the part of the digital image corresponding to this area of interest even if the lines of the postal addresses in the digital image are not oriented along an edge of the digital image as it is is the case for a postal object of the letter type.

An example of implementation of the method according to the invention is described in more detail and illustrated by the accompanying drawings.

Figure 1 is a simplified flowchart which illustrates the steps of the method according to the invention. Figure 2 is a schematic representation of a digital image of a package including a postal address.

FIG. 3 is a schematic representation of the image of a packet in FIG. 2 but partitioned into image blocks constituting thumbnails. FIG. 4 is a schematic representation of a group of thumbnails containing the information of the postal address of the package of FIG. 2.

FIG. 5 is a schematic representation of the area of interest of the image of FIG. 2 which is formed from the group of thumbnails of FIG. 4.

Figure 6 illustrates the principle of the Hough transformation on a thumbnail.

Figure 1 illustrates very schematically in the form of a flowchart part of a postal packet sorting process including steps of processing digital images of packets in order to automatically read an address on each postal packet which will be used in the sorting process to automatically direct the package in question to a corresponding sorting outlet. At 10, a package conveyed flat on a moving belt passes under a camera which captures a digital image of the package. Referring to FIG. 2 in which a digital IP image of a packet comprising an AP postal address has been represented schematically, it can be noted that in this image, the lines of the AP postal address are not written in parallel lines at the edge of the image. In general, such a digital image has been the subject of operations such as binarization, removal of the isolated black pixels, such operations being intended to reduce the size of the corresponding computer file and to remove the parasitic and useless data appearing. in the initial image. Thus, this image which was initially defined by pixels in shades of gray, will be reduced to black or white points, these points being able to symbolize according to the type of processing, a pixel, a grouping of pixels, or any other reduction of the pixels. of the initial image. In this way, the initial image can for example be reduced to a series of coordinates, each pair of coordinates defining a black point of the image.

According to the method of the invention, in step 20, the digital IP image (binarized and filtered) is divided or partitioned into identical image blocks called subsequently thumbnails. Referring to Figure 3, there is shown the image of Figure 1 partitioned into 72 IT thumbnails, each thumbnail being square.

The objective of this partitioning step is to identify in each thumbnail, in a preferred direction of the distribution of the black dots in each thumbnail. More particularly, the preferred direction of a thumbnail corresponds to the direction of one or more lines on which characters are inscribed in the thumbnail, and this preferred direction will be identified by the application of an algorithm to each thumbnail. For the result of the application of this algorithm to be relevant, it is important that the size of the thumbnails corresponds approximately to the size of a word, or to the size of several characters. Concretely, the thumbnails could for example be 2cm by 2cm squares on the package. An identification algorithm in a thumbnail of a preferred direction will be presented and detailed below. Thus, at the end of step 30, a preferred direction is known for each thumbnail, which generally corresponds to the direction of character lines of the thumbnail, or to a direction perpendicular to these lines. In 40, one or more groups of thumbnails are produced according to the criterion that the preferred directions of the thumbnails of a group are substantially parallel or substantially perpendicular, and that these thumbnails are contiguous. In Figure 4, there is shown the group of thumbnails Gl which has been identified with the method according to the invention in the IP image. As you can see, this group of thumbnails includes the AP address, but has not yet been shaped into a rectangle.

At 50, the areas of interest associated with each group of thumbnails from step 40 are produced. For a group of thumbnails, we will first calculate the most common or common preferred direction among the thumbnails of this group and associate this preferred direction with the group of thumbnails, then calculate the smaller rectangle comprising all the thumbnails of the group of thumbnails and having an edge parallel to the preferred direction of the group of thumbnails, so as to determine the area of interest corresponding to the group of thumbnails considered. Referring to Figures 4 and 5, we can see the area of interest Zl which has been established from the IP image, and which includes the recipient address entered in lines parallel to the lower edge of the rectangle formed by the area of interest Zl. More particularly, FIG. 5 is a representation of the area of interest Z1, as it has been extracted from the digital image IP, with the method according to the invention in step 50.

The area or areas of interest identified at 50 will then be examined at 60 by an automatic character recognition algorithm, according to the preferred direction associated with each area of interest, in order to extract from the packet the postal address sought.

A detection algorithm in a thumbnail of a preferred direction, as used in 30 could start with the application to this thumbnail of a polar Hough transformation. The Hough transformation, which is described in detail in the book “B. YU & AKJAIN Robust on fast squew detection algorithm for generic documents, Pattern Récognition Vol 29 n ° 10, page 1609”, allows to establish a polar table , from a thumbnail, this polar table giving for a certain number of lines of the thumbnail, the number of black dots contained in each of these lines. In such a polar table which has two inputs (r, θ), each pair (r, θ) corresponds to a straight line D (r, θ), the value of the polar table for a particular pair (r, θ) corresponding to the number of black dots that the line D (r, θ) contains. In Figure 6 there is shown a thumbnail IT which is a square of side C, in which we can see that the line D (r, θ) containing the black point N (x, y) is the line which passes through the point P (r, θ), and which is perpendicular to the vector OP (r, θ), O designating the origin of a coordinate system associated with the thumbnail. More particularly, the vector OP (r, θ) is a vector having for module r, and for polar angle θ. For a square thumbnail, on side C, provided with a marker centered on the lower left vertex of the image, the set of lines intercepting this square thumbnail is included in the set of lines D (r, θ) satisfying - C <r <CA / 2 and 0 ° <θ <180 °. Thus, in a thumbnail of 2cm by 2cm, as shown in Figure 6, for which we decide to discretize the data in 2mm steps for the rays, and in 10 ° steps for the angles, we can build a polar table having 24 columns, each column corresponding to a radius value, and 18 lines, each line corresponding to an angular value. This table, which will be initialized so as to be worth 0 for all the couples (r, θ), will then be progressively filled, starting from each black point N (x, y) of the thumbnail, with the relation r = x. cos (θ) + y.sin (θ) (*):

For a black point N (x, y), we will calculate the value taken by r for θ = 0 ° (with the relation (*)), we will round this value of r to 2mm, in accordance with the chosen discretization, to form a couple (r (0 °), 0 °). Then we will add 1 in the coordinates box (r (0 °), 0 °) of the polar table of the thumbnail. This operation must then be carried out for θ = 10 °, 20 °, 30 ° ... 170 °.

This sequence of operations must be carried out for each black point that the thumbnail includes to constitute the polar picture of the thumbnail. The polar array thus obtained then gives for each pair (r, θ), the number of black dots that the line D comprises (r, θ), this line being in this case a strip 2mm wide, taking into account the chosen discretization. To extract from this polar table the corresponding privileged direction of the thumbnail, we will calculate for each angle θ the sum s ² (θ), which is the sum of the squares of the terms of the entire line of the polar table, then choose the angle θ for which s ² (θ) is maximum.

The angle α corresponding to the preferred direction of the thumbnail is then equal to α = θ + 90 °. Indeed, if for example such a thumbnail consists of two black points aligned along a straight line oriented at 120 °, we have ² (30 °) = 2 ² = 4, and s ² (θ) = 1 ² +1 ² = 2 for θ ≠ 30 °, and therefore, the preferred direction provided by the application of this algorithm to this thumbnail will be α = 120 °.

As we can see, an algorithm based on the Hough transformation is efficient for detecting a privileged direction of the distribution of points in an image, but does not provide data on the location of points in the image, which is insufficient to identify areas of interest in the image of a postal package. By partitioning the image into thumbnails according to the method of the invention, before applying such an algorithm to each thumbnail, an analysis of the image is obtained which provides sufficient information to constitute areas of interest which can be quickly exploited by an automatic address recognition algorithm.

Claims

CLAIM

A method of processing a digital image (IP) of a postal object in order to automatically read a postal address (AP) on said object by application of an automatic address recognition algorithm on said image, characterized in that that it consists in dividing (20) the image into image blocks of identical size, in applying (30) to each image block a treatment based on a Hough transformation to identify in the image block considered a privileged direction of the distribution of the image points in said block considered, to group together (40) contiguous image blocks in the image and having a common privileged direction to define (50) a rectangular area of interest ( Zl) in which the grouping of said contiguous blocks having a common preferred direction is inscribed so that this rectangular area of interest has a longitudinal edge parallel to said common preferred direction, and to apply the algorithm me automatic recognition of addresses on said area of interest of the image in order to automatically read a postal address.